INDEX
Explanations
instances of the word "both" and the concept of shared experiences or duality
New Auto-Interp
Negative Logits
aba
-0.18
hin
-0.17
Äħ
-0.16
hana
-0.15
ABA
-0.14
ife
-0.14
hani
-0.14
thinkable
-0.14
hid
-0.14
stral
-0.13
POSITIVE LOGITS
//{{0.15
pant
0.15
ritz
0.15
æ©
0.15
otch
0.14
اÙĨت
0.14
Moor
0.14
cul
0.14
åĮ
0.14
ett
0.14
Activations Density 0.017%